clinical study amoxicillin/clavulanic acid for the...

Clinical StudyAmoxicillin/Clavulanic Acid for the Treatment ofOdontogenic Infections: A Randomised Study ComparingEfficacy and Tolerability versus Clindamycin

Archiel Launch Tancawan,1 Maria Noemi Pato,2 Khamiza Zainol Abidin,3

A. S. Mohd Asari,4 Tran Xuan Thong,5 Puja Kochhar,6 Chandra Muganurmath,7

Monique Twynholm,8 and Keith Barker9

1Tancawan Dental Clinic, Door 2 Cebu Coliseum Building, Leon Kilat Street, 6000 Cebu City, Philippines2East Avenue Medical Center, Dental Department, OPD Building, East Avenue, Diliman, 1100 Quezon City, Philippines3Klinik Pergigian Gunung Rapat, Unit Pakar Periodontik, Jalan Raja Dr. Nazrin Shah, 31350 Ipoh, Perak, Malaysia4Klinik Pergigian Cahaya Suria, Unit Periodontik, Tingkat 2, Bangunan Cahaya Suria, Jalan Tun Perak,50050 Kuala Lumpur, Malaysia5People’s Hospital 115, 527 Su Van Hanh Street, District 10, Ho Chi Minh City, Vietnam6GlaxoSmithKline Asia Pvt. Ltd., 252 Dr Annie Besant Road, Worli, Mumbai 400030, India7GlaxoSmithKline Pharmaceuticals Ltd., 5 Moore Drive, Research Triangle Park, NC 27709-3398, USA8GlaxoSmithKline Pharmaceuticals Ltd., Stockley Park West, 1-3 Iron Bridge Road, Heathrow, Uxbridge, Middlesex UB11 1BT, UK9GlaxoSmithKline Pharmaceuticals Ltd., GSK House, 980 Great West Road, Brentford, Middlesex TW8 9GS, UK

Correspondence should be addressed to Puja Kochhar; [email protected]

Received 27 April 2015; Accepted 6 July 2015

Academic Editor: Yoshitaka Hara

Copyright © 2015 Archiel Launch Tancawan et al. This is an open access article distributed under the Creative CommonsAttribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work isproperly cited.

Background. Treatment of odontogenic infections includes surgical drainage and adjunctive antibiotics. This study was designedto generate efficacy and safety data to support twice daily dosing of amoxicillin/clavulanic acid compared to clindamycin inodontogenic infections.Methods. This was a phase IV, randomised, observer blind study; 472 subjects were randomised to receiveamoxicillin/clavulanic acid (875mg/125mg BID, 𝑛 = 235) or clindamycin (150mg QID, 𝑛 = 237) for 5 or 7 days based on clinicalresponse. The primary endpoint was percentage of subjects achieving clinical success (composite measure of pain, swelling, fever,and additional antimicrobial therapy required) at the end of treatment. Results. The upper limit of two-sided 95% confidenceinterval for the treatment difference between the study arms (7.7%) was within protocol specified noninferiority margin of 10%,thus demonstrating noninferiority of amoxicillin/clavulanic acid to clindamycin. Secondary efficacy results showed a higher clinicalsuccess rate at Day 5 in the amoxicillin/clavulanic acid arm. Most adverse events (raised liver enzymes, diarrhoea, and headache)were similar across both arms and were of mild to moderate intensity. Conclusion. Amoxicillin/clavulanic acid was comparable toclindamycin in achieving clinical success (88.2% versus 89.7%) in acute odontogenic infections and the safety profile was consistentwith the known side effects of both drugs. Trial Registration. This trial is registered with Clinicaltrials.gov identifier: NCT02141217.

1. Introduction

Odontogenic infections are one of the most prevalent dis-eases worldwide and the principal reason for seeking dentalcare. Dental prescriptions account for nearly 7% to 11%of all common antibiotic prescriptions [1]. The commonest

emergency odontogenic infections are periapical abscess(25%), pericoronitis (11%), and periodontal abscess (7%) [2].

Odontogenic infections are mostly polymicrobial andfrequently encountered odontopathogens are Streptococcispp., Corynebacterium spp. and Staphylococcus spp., Pre-votella spp., Porphyromonas spp., Fusobacterium spp., and

Hindawi Publishing CorporationInternational Journal of DentistryVolume 2015, Article ID 472470, 9 pageshttp://dx.doi.org/10.1155/2015/472470

2 International Journal of Dentistry

Bacteroides spp. [2, 3]. Therapeutic success in odontogenicinfections is determined by the control of infection bysurgical debridement and/or antimicrobial therapy which isindicated when there are clear signs of systemic involvementsuch as pyrexia or lymphadenopathy [2]. The polymicro-bial nature of odontogenic infections necessitates the useof antibiotics active against both aerobic and anaerobicbacteria [4]. The antibiotics most commonly prescribed foracute dental abscesses are amoxicillin, penicillin, metron-idazole, and erythromycin with clindamycin as an alter-native in individuals allergic to the beta-lactam antibiotics[3].

Recently, published evidence suggests that penicillin andamoxicillin are being rendered increasingly less effectivebecause of beta-lactamase producing bacteria.More than halfof the Gram-negative anaerobic bacilli (including Prevotella,Porphyromonas, Bacteroides, and Fusobacterium spp.) arecapable of producing beta-lactamase leading to treatmentfailures in dental infections [5]. Studies have revealed thepresence of beta-lactamase producing species in 74–88% ofpatients with periodontitis [4]. Addition of a beta-lactamaseinhibitor such as clavulanic acid to amoxicillin (Augmentin)confers resistance to beta-lactamases thereby extending theantibiotic spectrum to anaerobes such as Prevotella spp.and Bacteroides spp. anaerobes and to Staphylococcus spp.[4]. Efficacy of amoxicillin/clavulanic acid in the treatmentof acute periapical abscess has been established in severalstudies [2, 4].

Clindamycin is a broad-spectrum antibiotic with activ-ity against aerobic, anaerobic bacteria including coverageagainst beta-lactamase producing pathogens [6]. Clinicaltrials have demonstrated the efficacy of clindamycin intreating odontogenic infections [7–11]. Use of clindamycin indental infections is based on careful patient selection in viewof reported cases of pseudomembranous colitis (a rare butserious consequence of clindamycin).

Despite published evidence evaluating different oral for-mulations of amoxicillin/clavulanic acid in dental infections,there is limited published data on the use of twice dailydosing of 875/125mg in odontogenic infections. Availableevidence suggests that twice daily dosing with 875/125mgamoxicillin/clavulanic acid results in a successful clinicaloutcome, better patient compliance, and less gastrointestinalupset, due to a reduction in the dose of clavulanic acid [1].Theaim of the current study was to assess the clinical efficacy andsafety of amoxicillin/clavulanic acid 875/125mg twice dailyversus clindamycin 150mg four times daily, for 5 or 7 days indental infections.

2. Methods

2.1. Study Design. AUG117044 was a phase IV, randomised,parallel group, comparative, observer blind study to evaluateefficacy, safety, and tolerability of amoxicillin/clavulanic acid(875mg/125mg) and clindamycin (150mg) in the treatmentof acute odontogenic infection with or without abscess. Thestudy was conducted in fifteen centres across four countrieswith four centres each in Malaysia, Philippines, and Vietnamand three inThailand.

The study protocol, the informed consent, and otherinformation that required preapproval were reviewed andapproved by a national, regional, or investigational centreethics committee or institutional review board, in accordancewith the International Conference on Harmonisation ofTechnical Requirements for Registration of Pharmaceuticalsfor Human Use (ICH) Good Clinical Practice (GCP) andapplicable country-specific requirements. Written informedconsent was obtained from each subject prior to the per-formance of any study-specific procedures. The study wasconducted in accordance with ICH GCP and all applicablesubject privacy requirements and the ethical principles thatare outlined in the Declaration of Helsinki 2008.

A total of 472 subjects were randomised in a 1 : 1 ratioin each of the treatment arms. The study included a one-day screening period followed by a treatment period of fivedays that could be extended to seven days based on clinicalresponse. Standard surgical intervention for odontogenicinfection was permitted only before commencing studytreatment. Eligible subjects were randomized on the day oftheir screening visit or within a day of screening. Efficacy andsafety evaluations were performed on Day 2, Day 5, and/orDay 7 (based on treatment duration) (Figure 1). Study treat-ment included amoxicillin/clavulanic acid (875mg/125mg)administered twice daily or clindamycin (150mg) adminis-tered four times daily alongwithmeals for 5 or 7 days. Clinicalefficacy (cure (cure was defined as complete resolution ofsigns and symptoms of infection present at baseline such thatno additional antimicrobial therapy was required), improve-ment (improvement was defined as resolution of fever (ifpresent at baseline) and >70% reduction in swelling and painand improvement in other signs and symptoms such that noadditional antimicrobial therapy was required), and failure(failure was defined as inability to improve the signs andsymptoms of infection after seven days of therapy so thatadditional antimicrobial therapy was required)) of the studytreatment was assessed based on the response shown by thesubjects on the Visual Analogue Scale (VAS) scores of painand swelling. Since these clinical efficacy parameters werebased on symptomatic relief, an optimal study design wouldhave been a double blind design. However, the differentdosage regimens and formulations of the study drugs pre-sented practical challenges for a double blind design. Basedon these considerations, the study was designed to be anobserver blind study with the investigator remaining blindedthroughout the study period. An unblinded study teammem-ber was appointed for the study for drug dispensing and drugaccountability and was also present during the clinic visits toensure that the investigators remained blinded to treatmentassignment. Adherence to the study design requirements wasessential and no protocol waivers or exemptions were allowedduring the study. This study did not require an independentdata safety monitoring board and no interim analysis wasperformed.

2.2. Study Population. Inclusion criteria: the study enrolledsubjects ≥18 years of age with a diagnosis of acute odonto-genic infections (periapical abscess, acute periodontitis, andpericoronitis) that required antibiotic therapy. Radiographic

International Journal of Dentistry 3

Day 5

Screeningperiod

Group 1

Group 2Randomisationand first dose

Day 2 Day 5 Day 7

Day 2

Two-arm, parallel, comparative, observer blind study

Informedconsent

Day 1

Day 0 Day 7

Treatment period

Day 0

of 5 to 7 ∗days

+ clavulanic acid (875mg/125mg)

150mg)

∗

Group 1: amoxicillin

Group 2: clindamycin (

Duration of treatment can be increased if deemed necessary by the investigator at the day 5 visit. In that case, clinicalevaluation will be done again at day 7.

Figure 1: Study schema. Note: the study had a 1-day screening period (Day 1 to Day 0) during which eligibility was assessed and laboratorytests were performed. Randomisation occurred within 24 hours of screening at baseline (Day 0). Further visits (Day 2, Day 5, and Day 7) werecalculated from the baseline/randomisation visit (Day 0).

evidence of odontogenic infection and dental pain on mas-tication were mandatory diagnostic criteria for enrollment.Exclusion criteria: subjects presenting with complicatedodontogenic infections (such as osteomyelitis, dentocuta-neous and dentoalveolar fistula, and facial-space swelling)or odontogenic infections secondary to traumatic injury orrequiring hospitalisation, aggressive intravenous antimicro-bial therapy, or local application of antimicrobials for thetreatment of odontogenic infection were excluded. Further,patients with other key exclusion criteria such as conditionsprone to infective endocarditis and those treated with sys-temic antibiotics within two weeks before the study of drugadministration or injectable long acting antibiotics adminis-tered four weeks prior to study of drug administration werealso excluded from study.

2.3. Study Variables. The primary objective of the study wasthe comparison of clinical efficacy of amoxicillin/clavulanicacid with clindamycin in subjects with acute odontogenicinfection with or without abscess. This efficacy endpoint wasbased on the percentage of subjects achieving clinical success(cure or improvement) at the end of treatment (Day 5 or 7).The secondary endpoints of the study included percentageof subjects achieving clinical success at Day 5 and change inthe VAS score for pain and swelling from baseline to Days 2,5, and 7. Safety assessments included monitoring of adverse

events (AEs) and serious adverse events (SAEs) from thestart of study treatment until the end of study treatment. Theantibiotic susceptibility of bacterial isolates obtained frompus specimens was recorded at baseline.

2.4. Statistical Analysis. A sample size of 205 evaluablesubjects in each of the treatment arms provided 90% powerto assess noninferiority for the primary endpoint. This wasbased on a noninferiority margin of 10%, assuming clinicalsuccess response rate in the comparator arm (clindamycin150mg) of 90% and a one-sided 𝛼 = 2.5%. Considering a15% drop-out rate, a final sample size of 236 subjects in eachstudy group was chosen. Thus total randomised subjects inthe studywere 472 for a 1 : 1 treatment allocation in each studygroup to get at least 205 evaluable subjects in each arm.

The per-protocol (PP) population was a subset of Intent-to-Treat (ITT) population that had a postbaseline clinical suc-cess response assessment and did not report major protocoldeviation(s). However, those subjects who discontinued fromthe study without any postbaseline assessment and wherethe reason for discontinuation was documented as “Lack ofEfficacy” or “Treatment Failure” were included in the PPpopulation with clinical success outcome treated as “ClinicalFailure.” For noninferiority analysis, the PP population usingobserved case (OC) method was treated as the primarydataset. The assessment of noninferiority for clinical success


response was based on two-sided 95% confidence interval(CI); the upper limit of two-sided 95% CI of the differenceof proportion between the two treatments of less than 0.10(10%) was set to conclude the noninferiority between thetreatment arms. The analysis of the Intent-to-Treat-Efficacy(ITT-E) population (all randomised subjects with at least onepostbaseline assessment of clinical success response) usingthe OC dataset was provided as a sensitivity analysis for theprimary endpoint and was also used to evaluate secondaryendpoints.The ITT population (all randomised subjects whoreceived at least one dose of study medication) was the safetydataset for the study.

The investigator’s judgment was considered decisive forthe assessment of clinical improvement in a subject. In theevent that the main signs and symptoms were cured orimproved (complete resolution of fever and >70% reductionin swelling and pain) and there was “no change” or “worsen-ing from baseline” in other signs and symptoms (suchas increased leucocyte count/tooth mobility/lymphadeno-pathy), the investigator’s opinion was sought as to whetheradditional antimicrobial therapy was required. Subjects thatrequired no additional antimicrobial therapy per the inves-tigators judgment were considered a “success” while thoserequiring additional antimicrobials were deemed as “fail-ures.” The analysis based on the investigator’s judgment ofclinical success or failure was considered as the primaryanalysis for testing of noninferiority between the treatmentgroups and the analysis excluding the investigator’s assess-ment for clinical success was presented as supportive analysis.

For sensitivity analysis, all subjects with an assessmentof cure or improvement (complete resolution of fever, >70%reduction in swelling and pain) but with “no change”or “worsening from baseline” in other signs and symp-toms (increased leucocyte count/tooth mobility/lymphaden-opathy) were considered as “Clinical Failures” irrespective ofthe clinical judgment of investigator.

Clinical laboratory assessments were summarized on ITTpopulation using OC approach for missing values. Liverfunction tests (LFTs) values were summarised after valueswere normalised. Any LFT parameter which was out ofreference range for that particular laboratory was consideredan adverse event (AE). However, AST, ALT, and alkalinephosphatase values >3 × upper limit of reference range(ULRR) and total bilirubin >1.5 × ULRR were considered tobe of potential clinical concern (PCC) as defined by sponsor.

3. Results

3.1. Participants. A total of 510 subjects were screened fora planned enrollment of 472 subjects. Amongst the 472randomised subjects, 235 (46.1%) subjects were randomisedto the amoxicillin/clavulanic acid arm and 237 (46.5%)to the clindamycin arm. However, a total of 236 sub-jects received amoxicillin/clavulanic acid and 235 subjectsreceived clindamycin (one subject randomised to the amoxi-cillin/clavulanic acid arm did not receive any study drug andtwo subjects randomised to the clindamycin arm incorrectlyreceived amoxicillin/clavulanic acid). A similar proportion ofenrolled subjects completed the study in both the treatment

arms (223 (94.9%) in the amoxicillin/clavulanic acid arm and229 (96.6%) in the clindamycin arm). A total of 11 (4.7%)subjects in the amoxicillin/clavulanic acid arm and 8 (3.4%)subjects in the clindamycin arm discontinued before studycompletion and themain reasons for discontinuationswere asfollows: AEs (one and two in amoxicillin/clavulanic acid andclindamycin arms, resp.), protocol noncompliance (one andfour in amoxicillin/clavulanic acid and clindamycin arms,resp.), and meeting the withdrawal criteria (nine and two inamoxicillin/clavulanic acid and clindamycin arms, resp.).

3.2. Baseline Characteristics. Subjects recruited in the studywere South East Asian in origin with similar age and sexdistribution across both treatment arms. Periapical abscesswas the predominant odontogenic infection across botharms (56.8% and 54.9% subjects in the amoxicillin/clavulanicacid and clindamycin arms, resp.). There was no significantdifference in baseline characteristics such as pain, swelling,radiographic evidence of dental infection, and medicalhistory/preexisting conditions between the treatment arms(Table 1).

3.3. Primary Efficacy Results. The primary efficacy analysisusing the PP population demonstrated that the clinicalefficacy of amoxicillin/clavulanic acid was noninferior toclindamycin, since the upper limit of two-sided 95% CI waswithin the protocol specified noninferiority margin of 10%.

The percentage of subjects achieving clinical successusing the primary analysis population was 88.2% (95% CI:83.0%, 92.2%) in the amoxicillin/clavulanic acid arm and89.7% (95% CI: 84.6%, 93.5%) in the clindamycin arm.The treatment difference between the treatment arms was1.5% (95% CI: −4.7%, 7.7%) using Miettinen and Nurminenmethod, 1.5% (95% CI: −4.9%, 8.0%) using Farrington andManning method, and 1.5% (95% CI: −4.5%, 7.6%) usinga two-sample proportion test. Since the upper limit of thetwo-sided 95% CI for between-group percentages differenceswas less than the prespecified noninferiority margin of 10%,noninferiority of amoxicillin/clavulanic acid to clindamycinwith respect to clinical success was demonstrated. This resultwas also corroborated by sensitivity analysis using the ITT-Epopulation (Table 2).

3.4. Secondary Efficacy Results. A slightly higher percent-age of subjects achieved clinical success in the amoxi-cillin/clavulanic acid arm by Day 5 [76.8% (95% CI: 70.7%,82.2%)] than the clindamycin arm [69.1% (95% CI 62.7%,75.0%)] possibly indicating a faster response in subjectsreceiving amoxicillin/clavulanic acid arm as compared tosubjects who received clindamycin.

The least square mean change in the VAS score for pain(using ITT-E dataset with OC approach) was maximum atDay 7 (6.38 and 6.34 in the amoxicillin/clavulanic acid andclindamycin arms, resp.) compared to Day 5 (5.49 and 5.38in the treatment arms, resp.) and Day 2 (3.34 and 3.07, resp.)and it was similar between the treatment arms at each timepoint. Similar results were also noted for swelling (Table 3). Asummary of VAS by visit and treatment arms demonstratedthat higher mean percentage reduction in pain by Day 2


Table 1: Summary of demographics and baseline characteristics (ITT population).

Characteristic StatisticRandomised treatment arm

Amoxicillin/clavulanic acid(𝑁 = 234)

Clindamycin(𝑁 = 237)

GenderMale 𝑛 (%) 99 (42.3%) 94 (39.7%)Female 𝑛 (%) 135 (57.7%) 143 (60.3%)

Geographic ancestryAsian: Central/South Asian heritage 𝑛 (%) 0 1 (0.4%)Asian: East Asian heritage 𝑛 (%) 1 (0.4%) 4 (1.7%)Asian: South East Asian heritage 𝑛 (%) 233 (99.6%) 232 (97.9%)

Age (years)

𝑁 (missing) 234 (0) 237 (0)Mean (SD) 33.1 (12.8) 32.6 (12.0)Median 29.2 28.9

(Min, max) (18.0, 74.8) (18.1, 69.0)Type of odontogenic infection

Periapical abscess 𝑛 (%) 133 (56.8%) 130 (54.9%)Acute periodontitis 𝑛 (%) 40 (17.1%) 37 (15.6%)Pericoronitis 𝑛 (%) 62 (26.5%) 73 (30.8%)

Diagnostic criteria fulfilledDental pain which is increased on mastication 𝑛 (%) 234 (100.0%) 237 (100.0%)Swelling on alveolar mucosa 𝑛 (%) 226 (96.6%) 227 (95.8%)Redness over involved region 𝑛 (%) 215 (91.9%) 219 (92.4%)Increased tooth mobility 𝑛 (%) 82 (35.0%) 75 (31.6%)Fever 𝑛 (%) 5 (2.1%) 5 (2.1%)Malaise 𝑛 (%) 17 (7.3%) 21 (8.9%)Cervical lymphadenopathy 𝑛 (%) 27 (11.5%) 26 (11.0%)Elevated leucocyte count 𝑛 (%) 37 (15.8%) 37 (15.6%)

Baseline VAS score

Dental pain


(Min, max) (1.0, 10.0) (0.5, 10.0)

Swelling


(Min, max) (0.6, 10.0) (0.1, 10.0)Surgical intervention required prior to study treatment 𝑛 (%) 55 (23.5%) 55 (23.2%)Percentage is calculated based on number of subjects in ITT population for each treatment arm and by actual randomised arm. As per randomisation, therewere 234 subjects in amoxicillin/clavulanic acid arm and 237 in clindamycin arm but due to wrong randomisation process, two subjects were administeredamoxicillin/clavulanic acid instead of clindamycin. Hence as per actual treatment received, there were 236 subjects in amoxicillin/clavulanic acid arm and 235subjects in clindamycin arm.

was achieved in the amoxicillin/clavulanic acid arm (49.5%)compared with (45.6%) the clindamycin arm. Similarly, ahigher mean percentage reduction in swelling by Day 2was achieved in the amoxicillin/clavulanic acid arm (43.6%)compared with the clindamycin arm (39.6%).

3.5. Microbiological Results. Pus specimens were obtainedin 58 subjects who consented to microbiological sampling

including two who were screen failures (25 in the amoxi-cillin/clavulanic acid arm and 31 in the clindamycin arm; the2 screen failures were excluded). A total of 61 isolates wereobtained from 56 samples, 26 in the amoxicillin/clavulanicacid arm and 35 in the clindamycin arm. Organisms isolatedin both the treatment arms were similar and predominantlyviridans streptococci group ((𝑛 = 24) including Streptococcusoralis, Streptococcus mitis, and Streptococcus parasanguinis),


Table 2: Primary efficacy endpoint: clinical success outcome at the end of the study.

Population Clindamycin[% (95% CI)]

Amoxicillin/clavulanicacid [% (95% CI)]

Treatment differenceMiettinen and

Nurminen method(primary) [% (95% CI)]

Farrington and Manningmethod [% (95% CI)]

Two-sampleproportion test [%

(95% CI)]

PP 89.7%(84.6%, 93.5%) 88.2% (83.0%, 92.2%) 1.5% ( 95 CI: −4.7%,

7.7%)1.5%

(−4.9%, 8.0%) 1.5% (−4.5%, 7.6%)

ITT-E 86.4%(81.3%, 90.5%) 85.5% (80.3%, 89.8%) 0.9% (−5.5%, 7.3%) 0.9%

(−5.6%, 7.4%) 0.9% (−5.5%, 7.2%)

ITT(randomisedtreatment arm)

85.7%(80.5%, 89.9%) 83.3% (77.9%, 87.9%) 2.3% (−4.3%, 9.0%) 2.3%

(−4.4%, 9.0%) 2.3% (−4.2%, 8.9%)

Table 3: Secondary endpoint: change in VAS scores for pain and swelling from baseline.

Secondary endpoint: change in VAS from baseline

Assessment Day 𝑗 StatisticAmoxicillin/clavulanic

acid(𝑁 = 228)


Treatmentdifference

Change in pain at Day 𝑗from baseline

𝑁 (missing) 227 (1) 233 (2) —Day 2 Least square mean 3.34 3.07 0.27

Two-sided 95% CI (LCL, UCL) (3.08, 3.61) (2.81, 3.33) (−0.10, 0.64)𝑁 (missing) 219 (9) 228 (7) —

Day 5 Least square mean 5.49 5.38 0.11Two-sided 95% CI (LCL, UCL) (5.27, 5.71) (5.16, 5.60) (−0.20, 0.42)

𝑁 (missing) 57 (0) 71 (0) —Day 7∗ Least square mean 6.38 6.34 0.04

Two-sided 95% CI (LCL, UCL) (6.02, 6.74) (6.02, 6.66) (−0.44, 0.53)

Change in swelling at Day 𝑗from baseline

𝑁 (missing) 219 (2) 225 (1) —Day 2 Least square mean 1.92 1.61 0.31

Two-sided 95% CI (LCL, UCL) (1.72, 2.11) (1.42, 1.80) (0.04, 0.57)𝑁 (missing) 214 (7) 223 (3) —

Day 5 Least square mean 3.68 3.60 0.08Two-sided 95% CI (LCL, UCL) (3.51, 3.85) (3.43, 3.76) (−0.16, 0.32)

𝑁 (missing) 55 (0) 68 (0) —Day 7∗ Least square mean 4.21 4.61 −0.39

Two-sided 95% CI (LCL, UCL) (3.94, 4.49) (4.36, 4.86) (−0.77, −0.02)Note: change from baseline in VAS for assessment of pain and swelling was analysed using a mixed model for repeated measures (MMRM) with restrictedmaximum likelihood and an unstructured covariance matrix.The baseline VAS score was used as a covariate. Treatment groups and nominal days (visits) wereconsidered as fixed effects and interaction effect was considered between treatment and visit (days). Data for “Day 7” is summarised for subjects who continuedthe study up to Day 7. OC method was used for missing values where the missing value was kept as missing except for early withdrawal.“𝑗” indicates Day 2, or Day 5, or Day 7 as applicable.LCL: lower confidence limit, UCL: upper confidence limit.∗means the interaction effect between treatment and visits (days).

Enterobacter spp. (𝑛 = 10), Klebsiella spp. (𝑛 = 9), Pseudo-monas spp. (𝑛 = 7), and Staphylococcus spp. (𝑛 = 5). CLSIbreakpoints are not uniformly available for all isolates forboth study drugswhich posed a challenge in providingmean-ingful interpretation of susceptibility data.

3.6. Safety Results. A total of 243 treatment emergent AEs(TEAEs) were reported in 123 subjects in the amoxi-cillin/clavulanic acid arm and 236 events were reported in

124 subjects in the clindamycin arm. The most frequentlyobserved TEAEs with frequency ≥3% were abdominal dis-comfort, raised liver enzymes (AST, ALT, and bilirubin), diar-rhoea, dizziness, headache, increased appetite, and somno-lence (Table 4). Generally the incidence of TEAEswas similarbetween treatment arms, except for diarrhoea and headachewhich were reported in slightly more patients in the clin-damycin arm. Most AEs were of mild to moderate intensity.The incidence of drug related TEAEs in both the treatment


Table 4: Summary of commonly observed treatment emergent AEs(≥3% in either of the arms).

MedDRA preferred termAmoxicillin/clavulanic

acid(𝑁 = 236)


Total number oftreatment emergent AEs 243 236

Subjects whoexperienced at least oneAE

123 (52.1%) 124 (52.8%)

Abdominal discomfort 11 (4.7%) 7 (3.0%)Alanineaminotransferaseincreased

26 (11.0%) 24 (10.2%)

Aspartateaminotransferaseincreased

24 (10.2%) 20 (8.5%)

Blood bilirubinincreased 12 (5.1%) 13 (5.5%)

Diarrhoea 19 (8.1%) 28 (11.9%)Dizziness 18 (7.6%) 14 (6.0%)Headache 8 (3.4%) 14 (6.0%)Increased appetite 20 (8.5%) 15 (6.4%)Nausea 7 (3.0%) 4 (1.7%)Somnolence 19 (8.1%) 17 (7.2%)

arms was similar (165 AEs reported in 93 (39.4%) subjectsand 171 AEs reported in 97 (41.3%) subjects in the amoxi-cillin/clavulanic acid and clindamycin arms, resp.). The mostfrequently reported drug related TEAEs were gastrointestinaldisorders including abdominal discomfort, diarrhoea, nauseaand vomiting, abnormal LFTs, increased appetite, somno-lence, dizziness, and headache. Most of the related TEAEswere mild to moderate in intensity except for six eventsof severe intensity. These were elevated ALT, headache, andvomiting reported in the amoxicillin/clavulanic acid armand burning sensation, hypertension, and hypersomnia inthe clindamycin arm. All events had resolved by the endof the study except hypersomnia. A total of 89 subjects inthe amoxicillin/clavulanic acid arm and 76 subjects in theclindamycin arm had AEs that remained ongoing at the endof the study. Increased LFTwas the predominant ongoing AEand was present in 46 subjects in the amoxicillin/clavulanicacid arm and 40 subjects in clindamycin arm, respectively.The probable reason for the ongoing AEs could be the shortduration of the study (7 to 8 days) and the lack of a plannedfollow-up visit after study treatment. No subjects in the studyshowed shift in ALT and AST from normal at baseline (withrespect to local laboratory reference range) to PCC range(as defined in statistical analysis section) at end of the studywhereas three subjects each in both the study arms showedshift in total bilirubin to PCC range at the end of the study.However, since these subjects only had increased bilirubinwith no increase in ALT or AST, these were not of clinicalconcern. Four subjects in the clindamycin arm with high

LFT parameter (one subject each for ALT and AST and twosubjects for total bilirubin) at baseline remained in the PCCrange at the end of the study (Figure 2). No events of SAE ordeath were reported in the study.

4. Discussion

The primary treatment in acute odontogenic infections issurgical drainage while antibiotics are an adjunct in patientsshowing signs of systemic involvement [12]. The polymi-crobial component of odontogenic infection necessitates theuse of antibiotics that are active against both aerobic andanaerobic bacteria and therefore are recommended [4]. Thecurrent study was aimed at comparing the clinical efficacy ofamoxicillin/clavulanic acid (875mg/125mg) to clindamycin(150mg) in subjects with acute odontogenic infections.

Amoxicillin/clavulanic acid (875mg/125mg) was admin-istered twice daily for 5–7 days and was found to be noninfe-rior or “comparable” to clindamycin (150mg) administeredfour times daily. The overall clinical success seen withamoxicillin/clavulanic acid in the study (88.2% (95% CI:83.0%, 92.2%)) was similar to results seen in other publishedstudies. Success rates of 87% with amoxicillin/clavulanicacid 1 g twice daily and 96% with amoxicillin/clavulanicacid 625mg thrice daily have been previously reported inother studies [13, 14]. A higher percentage of pain andswelling reduction (49.5% and 43.6%) was achieved in theamoxicillin/clavulanic acid arm compared to the clindamycinarm (45.6% and 39.6%) after two days of treatment. In asimilar study aimed at demonstrating possible differencesin the severity of symptoms after the use of amoxicillinand amoxicillin/clavulanic acid in dental ailments, pain wasfound to be less acute with amoxicillin/clavulanic acid [15].Thus, these results indicate that amoxicillin/clavulanic acidgiven twice daily for the treatment of odontogenic infectionserves as an appropriate treatment option with the potentialadvantage of an early clinical response.

Safety of the subjects, assessed throughout the study,showed that the overall incidences of TEAEs were similaracross both the treatment arms. These were mainly eventssuch as abdominal discomfort, diarrhoea, and raised liverenzymes. As per the available safety information of amox-icillin/clavulanic acid 875/125mg, nausea and diarrhoea arecommonly reported events [16]. Diarrhoea was seen in 8.1%of subjects in amoxicillin/clavulanic acid arm (compared to11.9% in the clindamycin arm) and this is consistent withthe known pharmacological effects of the drug and with theglobal prescribing information. Raised liver enzymes are aknown but uncommon side effect (≥1/1,000 to <1/100) ofamoxicillin/clavulanic acid. In the current study, a total of 34subjects (14.4%) in the amoxicillin/clavulanic acid arm hadraised liver enzymes posttreatment compared to 33 subjects(14.04%) in the clindamycin arm that were assessed by theinvestigator as related to the study drug. However, most ofthese events were of mild to moderate intensity and noneof the subjects were considered to have any LFT values thatwere of clinical concern. Pseudomembranous colitis is a rarebut serious side effect of both clindamycin and amoxicillin-clavulanic acid. However, literature evidence suggests that the


5

4

3

2

1

0

543210 543210 543210

End

of st

udy

norm

alise

d va

lue

ALT AST Total bilirubin

Baseline normalised value

Actual treatment arm

ClindamycinAmoxicillin + clavulanic acid

Figure 2: Scatter plot of LFT with respect to baseline and end of study by treatment. Normalised value = laboratory value/upper limit ofnormal reference range of the respective local laboratory.

incidence is particularly low when these antibiotics are givenin outpatient care settings (6.7 cases/100,000 antibiotic expo-sures). In dental infections there is not much difference in theincidence of C. difficile colitis between the two drugs [6]. Inthe current study there were no cases of pseudomembranouscolitis with either of the study drugs. There were no SAEs ordeaths reported in this study.

One of the main limitations of the study was the useof an outcome measure based on a subjective score (VASscore) to assess pain and swelling to derive the compositeclinical outcome. However, to overcome the subjectivity ofthe VAS score and to reflect real world practice, only subjectsdemonstrating >70% reduction in pain and swelling wereconsidered for calculating clinical success response. Anotherlimitation was not having a planned follow-up visit for thesubjects after the end of the study visits. As a result, ongoingAEs typically key laboratory parameters such as liver enzymescould not be followed until resolution; however, most weremild and transient in nature.

5. Conclusions

Amoxicillin/clavulanic acid (875mg/125mg) administeredtwice daily was found to be comparable to clindamycin(150mg) administered four times daily in achieving clinicalsuccess in acute odontogenic infections with or withoutabscess. It was also found to be well tolerated with a safetyprofile consistent with the known pharmacologic effects ofamoxicillin/clavulanic acid and with that described in theglobal prescribing information.

Conflict of Interests

Archiel Launch Tancawan, Maria Noemi Pato, KhamizaZainol Abidin, A. S. Mohd Asari, and Tran Xuan Thong

declare no conflict of interests regarding the publication ofthis paper. Puja Kochhar, Chandra Muganurmath, MoniqueTwynholm, andKeith Barker are employees of GlaxoSmithK-line and hold stocks/shares in GlaxoSmithKline. ArchielLaunch Tancawan, Maria Noemi Pato, Khamiza ZainolAbidin, A. S. Mohd Asari, and Tran Xuan Thong wereprincipal investigators in AUG117044 study.

Acknowledgments

Nooraini Bt Osman (Putrajaya Dental Clinic, Malaysia),Christopher Vincent (Selayang Hospital, Malaysia), HaideeDaphnee Digma (Southern Philippines Medical Center,Philippines), Margarita Oasin (Rizal Medical Center, Philip-pines), Ngo Thi Quynh Lan (University of Medicine andPharmacy, Vietnam), Huynh Dai Hai (Odonto-Maxillo-Facial Hospital, Vietnam), Dao Thi Dung (Vietnam-CubaFriendship, Hanoi), Tanakrit Noppakunwijai (Tanarat FortHospital,Thailand), Prisana Pripatnanont (Prince of SongklaUniversity, Thailand), and Nipon Chaisrisookumporn (Lam-pang Regional Hospital, Thailand) were principal investi-gators. Gandhali Paranjape and Kavitha Rangappa (Diag-noSearch Life Sciences Pvt. Ltd.) were responsible foreditorial assistance; Boshi Mohala (GlaxoSmithKline) wasresponsible for safety review; JieDing (GlaxoSmithKline) andVarsha Parulekar (DiagnoSearch Life Sciences Pvt. Ltd.) wereresponsible for statistical support. Study was funded by Glax-oSmithKline Pharmaceuticals Ltd. and editorial assistancewas provided by DiagnoSearch Life Sciences Pvt. Ltd.

References

[1] N. S. Dar-Odeh, O. Abu-Hammad, M. K. Al-Omiri, A. S.Khraisat, andA. A. Shehabi, “Antibiotic prescribing practices bydentists: a review,”Therapeutics and Clinical Risk Management,vol. 6, pp. 301–306, 2010.


[2] R. Lopez-Pıriz, L. Aguilar, and M. J. Gimenez, “Management ofodontogenic infection of pulpal and periodontal origin,”Medi-cinaOral, Patologıa Oral y Cirugıa Bucal, vol. 12, no. 2, pp. E154–E159, 2007.

[3] D. Robertson and A. J. Smith, “The microbiology of the acutedental abscess,” Journal of Medical Microbiology, vol. 58, no. 2,pp. 155–162, 2009.

[4] A. B. Martınez, J. M. A. Urızar, A. B. Fenoll et al., “Consensusstatement on antimicrobial treatment of odontogenic bacterialinfections,” Medicina Oral, Patologia Oral y Cirugia Bucal, vol.9, no. 5, pp. 363–376, 2004.

[5] I. Brook, E. H. Frazier, and M. E. Gher, “Aerobic and anaerobicmicrobiology of periapical abscess,” Oral Microbiology andImmunology, vol. 6, no. 2, pp. 123–125, 1991.

[6] I. Brook, M. A. O. Lewis, G. K. B. Sandor, M. Jeffcoat, L. P.Samaranayake, and J. V. Rojas, “Clindamycin in dentistry: morethan just effective prophylaxis for endocarditis?,” Oral Surgery,Oral Medicine, Oral Pathology, Oral Radiology and Endodon-tology, vol. 100, no. 5, pp. 550–558, 2005.

[7] C. Walker and J. Gordon, “The effect of clindamycin on themicrobiota associated with refractory periodontitis,” Journal ofPeriodontology, vol. 61, no. 11, pp. 692–698, 1990.

[8] W. C. Gilmore, N. V. Jacobus, S. L. Gorbach, H. C. Doku, andF. P. Tally, “A prospective double-blind evaluation of penicillinversus clindamycin in the treatment of odontogenic infections,”Journal of Oral and Maxillofacial Surgery, vol. 46, no. 12, pp.1065–1070, 1988.

[9] L. Von Konow, P. A. Kondell, C. E. Nord, and A. Heimdahl,“Clindamycin versus phenoxymethylpenicillin in the treatmentof acute orofacial infections,” European Journal of ClinicalMicrobiology and Infectious Diseases, vol. 11, no. 12, pp. 1129–1135, 1992.

[10] S. Mangundjaja and K. Hardjawinata, “Clindamycin versusampicillin in the treatment of odontogenic infections,” ClinicalTherapeutics, vol. 12, no. 3, pp. 242–249, 1990.

[11] J. Gordon, C. Walker, C. Hovliaras, and S. Socransky, “Efficacyof clindamycin hydrochloride in refractory periodontitis: 24-month results,” Journal of Periodontology, vol. 61, no. 11, pp. 686–691, 1990.

[12] S. J. Ellison, “An outcome audit of three day antimicrobialprescribing for the acute dentoalveolar abscess,” British DentalJournal, vol. 211, no. 12, pp. 591–594, 2011.

[13] B. J. Al-Selivany, N. A. Al-Derzi, and S. Y. Agha, “Dental infec-tions: clinical and microbiological evaluation of responsivenessto twice daily amoxicillin-clavulanic acid (amoxiclave),” JordanMedical Journal, vol. 44, no. 3, pp. 305–312, 2010.

[14] C. F. Adriaenssen, “Comparison of the efficacy, safety and toler-ability of azithromycin and co-amoxiclav in the treatment ofacute periapical abscesses,” Journal of International MedicalResearch, vol. 26, no. 5, pp. 257–265, 1998.

[15] N. Sulejmanagic, H. Sulejmanagic, Z. Ljutovic, D. Salihagic, andM. Sijercic, “Combined application of amoxicillin and clavu-lanic acid after oral surgical interventions,” Bosnian Journal ofBasic Medical Sciences, vol. 5, no. 1, pp. 61–68, 2005.

[16] AUGMENTIN 875/125MG, “Amoxicillin/clavulanic acid,” Pro-duct Information, 2013.

Submit your manuscripts athttp://www.hindawi.com

Hindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Oral OncologyJournal of

DentistryInternational Journal of



International Journal of

Biomaterials


BioMed Research International


Case Reports in Dentistry


Oral ImplantsJournal of


Anesthesiology Research and Practice


Radiology Research and Practice

Environmental and Public Health

Journal of


The Scientific World JournalHindawi Publishing Corporation http://www.hindawi.com Volume 2014


Dental SurgeryJournal of

Drug DeliveryJournal of



Oral DiseasesJournal of


Computational and Mathematical Methods in Medicine

ScientificaHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

PainResearch and TreatmentHindawi Publishing Corporationhttp://www.hindawi.com Volume 2014

Preventive MedicineAdvances in


EndocrinologyInternational Journal of



OrthopedicsAdvances in

clinical study amoxicillin/clavulanic acid for the...

Documents